ORIGINAL PAPER

Geological challenges in constructing the proposed Geba dam site,northern Ethiopia

Gebremedhin Berhane • Kristine Walraevens

Received: 28 February 2012 / Accepted: 26 April 2012 / Published online: 6 October 2013

� Springer-Verlag Berlin Heidelberg 2013

Abstract It is proposed to construct a dam across the

Geba River, Ethiopia. The paper reports the engineering

geological investigations undertaken, including mapping,

discontinuity surveys, core drilling, water absorption test-

ing and sampling for laboratory tests. The complexity of

the site, with limestones and interbedded limestone-shale

horizons, is indicated by the variability of the RQD and

Lugeon values. Of the 63 tests undertaken, some two-thirds

had Lugeon values implying grouting was necessary. Fol-

lowing removal and replacement of the alluvial deposits in

the central area, a grout curtain including two to three rows

of grouting holes was recommended to a depth of 100 m

for the left abutment, 35 m for the central foundation and

60 m for the right abutment.

Keywords Dam site � Engineering geological

mapping � Geba River � Lugeon test � Northern

Ethiopia

Resume La construction d’un barrage sur le fleuve Geba,

en Ethiopie, est projetee. L’article presente les reconnais-

sances geologiques et geotechniques realisees, comprenant

une cartographie, des levers de discontinuites, des carot-

tages, des essais d’absorption d’eau et des echantillonnages

pour les essais de laboratoire. La complexite du site, avec

des formations calcaires et des alternances de schistes et

calcaires, se traduit par des valeurs de RQD et des resultats

d’essais Lugeon tres divers. Sur les 63 essais Lugeon rea-

lises, environ les deux tiers presentent des valeurs impli-

quant des travaux d’injection. Apres l’enlevement et le

remplacement des depots alluviaux dans la zone centrale de

la fondation, un ecran d’injection constitue de deux a trois

rangees de forages d’injection a ete recommande jusqu’a

une profondeur de 100 m pour l’appui de rive gauche,

35 m pour la partie centrale de la fondation et 60 m pour

l’appui de rive droite.

Mots cles Site de barrage � Cartographie de geologie

de l’ingenieur �Riviere Geba � Essai Lugeon � Ethiopie

du Nord

Introduction

One of the most important environmental issues faced by

various countries is the lack of an adequate water supply. It

has been estimated that nearly two-thirds of nations

worldwide will experience water shortages by the year

2025 (United Nations Environment Programme 2002).

Koutsoyiannis (2011) has indicated that, due to the growth

of population and average per capita water use, the amount

of fresh water withdrawn globally each year has increased

from 579 km3 in 1900 to 3,973 km3 in 2000 and demand is

projected to rise further to 5,235 km3 by 2025. He con-

cluded that more dams are needed worldwide to meet

increased water and food supply needs.

According to the United Nations Population Division

(2002), by the year 2030, 56 % of people in developing

counties will reside in urban areas. In Ethiopia, some 17 %

of the population was residing in urban centres in 2010;

this number is projected to reach 21 % in 2025 (United

G. Berhane (&)

Department of Earth Sciences, College of Natural and

Computational Sciences, Mekelle University, P.O. Box 1202,

Mekelle, Ethiopia

e-mail: gmedhin_berhane@yahoo.com

G. Berhane � K. Walraevens

Laboratory for Applied Geology and Hydrogeology, Ghent

University, Krijgslaan 281-S8, 9000 Ghent, Belgium

e-mail: Kristine.Walraevens@UGent.be

123

Bull Eng Geol Environ (2013) 72:339–352

DOI 10.1007/s10064-013-0480-9

Nations Population Division 2010), presaging water

shortages. Climate changes and the continuing trend of

population migration to cities, mainly in developing

countries, will aggravate the problem. In Eastern Africa, a

30 % improvement in access to piped water sources over

the last 30 years has been reported, more than half of

which relates to urban areas (Lee and Schwab 2005).

However, many cities in this region are still suffering from

a shortage of potable water and from water-borne diseases.

In Ethiopia efforts to alleviate the problem are being

made by both the federal government and local adminis-

trative bodies, including the proposed Geba dam to be built

across the Geba River, 25 km northwest of Mekelle city

(Fig. 1). The dam will have a crest length of 1,000 m, a

height of 80 m and a reservoir capacity of 350 million

cubic metres. It is designed both to supply water to the city

and surrounding populations and to regulate the flow of the

river. Currently, Mekelle city is supplied with groundwater

mainly from the Aynalem well field, some 5 km to the

south east. In 1998, 11 wells were drilled and pump tested

for 72 h. It was concluded that total output was some 220 l/s,

with each well providing between 15 and 30 l/s. However,

since 2002, the water level has declined continuously

(Water Works Design and Supervision Enterprise,

WWDSE 2007).

Drought in Ethiopia is a frequently recurring phenom-

enon and its distribution and frequency have increased in

recent years (Walraevens et al. 2009). In addition to

excessive abstraction, climatic variability might be con-

tributing to the decline in the water level in the well field.

Nothing better captures the enormity of the scarcity of

water than the fact that water is only provided to residents

for a limited number of days per week. Clearly, alternative

ways to meet the rapidly growing water demand of the city

must be found. In the last two decades the government has

initiated the construction of micro-dam reservoirs and other

water-harvesting structures for different purposes but the

proposed Geba dam (Fig. 1) is a very important project.

Numerous dams, all over the world, are affected by

leakage when filled (Mozafari et al. 2011). The hydraulic

and mechanical properties of rock masses are the most

important parameters in the design and construction of

dams (Gurocak and Alemdag 2011) and other water-har-

vesting and retaining structures. The permeability of rock

masses in general and related to dam design and con-

struction has been studied by numerous researchers (Bon-

acci and Roje-Bonacci 2008; Foyo et al. 1997, 2005;

Goodman et al. 1965; Heuer 1995; Izharul and Hashmi

1983; Kiraly 1969, 1978; Ozsan and Karpuz 1996; Snow

1968; Uromeihy and Farrokhi 2011; Yamaguchi et al.

1997). The Lugeon test is widely used to estimate the

average hydraulic conductivity of rock masses (Camilo

Quinones-Rozo 2010) and is often considered both as the

most important parameter in determining the critical per-

meability and as a criterion to determine the necessity of

rock mass grouting (Sharghi et al. 2010).

The permeability of naturally occurring geological strata

is important for foundation and underground construction,

hydraulic structures and groundwater, oil and gas exploi-

tation (Angulo et al. 2011; Morrow 2000). In-situ

Fig. 1 Location map of Geba dam site and its environs

340 G. Berhane, K. Walraevens

123

permeability tests of soil and rock usually provide a more

accurate determination of permeability than laboratory

tests (Hamm et al. 2007; Mollah and Sayed 1995). How-

ever, due to the heterogeneity of the hydrogeological

characteristics of rock masses and the anisotropy arising

from discontinuities, permeability values determined by

in situ tests in a limited area may not reflect the real per-

meability of the rock mass at a project site scale. The

Lugeon (packer) test is the most commonly used in situ test

(Lugeon 1933), but in situ tests in drill holes only provide

information on the permeability of the strata immediately

adjacent to the borehole (Gurocak and Alemdag 2011).

Detailed analysis and correlation with the geology of the

site are necessary to complement the data from drill holes.

The geology of northern Ethiopia in general and that of

the Mekelle Outlier in particular has been described by

such authors as Beyth (1971, 1972), Levitte (1970) and

Wolela (2008). However, with the exception of a few

localized and project-specific works (Berhane 2010a, b;

Berhane and Ayenew 2010), almost no studies have been

devoted to the engineering geological or geotechnical

aspects of the Mesozoic sedimentary rocks of the Mekelle

Outlier.

About 70 micro-dam reservoirs have been constructed

during the last two decades, mainly located in the sedi-

mentary basin of the Mekelle Outlier. However, due to

technical and operational problems most of the micro-dams

suffer significant leakage (Abdulkadir 2009; Berhane

2010a; Desta 2005; Gonzalez-Quijano 2006; Haregeweyn

et al. 2005; Nedaw and Walraevens 2009), largely due to

the inadequacy of the initial hydrogeological, engineering

geological and geotechnical investigations.

The objective of this study was to evaluate and char-

acterize the hydraulic conductivity of the rock masses at

the site of the proposed Geba dam, and to check whether

grouting is necessary and practical in the abutments and

central foundation.

Background geology

The first recorded geological work in the northern prov-

inces of Ethiopia was carried out by Blanford (1870, cited

in Beyth 1971), who divided the Trap Volcanics of the

Ethiopian highlands into two units: a lower entirely basaltic

Ashangi Series and an upper Magdala Series, containing

many intercalations of trachyte. Dainelli and Marinelli

(1912) and Merla and Minucci (1943, as cited in Beyth

1971) proposed transgression–regression phenomena to

explain the sedimentary history of the whole of the Horn of

Africa, including Ethiopia. In 1970, Levitte studied the

geology of the Mekelle area and divided the rocks into four

major units: Basement Complex, Palaeozoic—Mesozoic

Sedimentary Sequence, Cenozoic Trap Volcanics and

Sediments of the Ethiopian Rift (Fig. 2). Beyth (1972)

undertook detailed mapping of the northern Ethiopian

provinces (Central and Western Tigray regions) and sug-

gested the formation of the Mekelle Outlier began in either

the Ordovician or Carboniferous period and probably

ended in the Lower Cretaceous before the eruption of the

Trap Volcanics.

According to Wolela (2008), about 33 % of the surface

area of Ethiopia is covered by sedimentary rocks in five

major basins (the Ogaden Basin, the Blue Nile Basin, the

Gambela Basin, the Southern Rift Basin and the Mekelle

Outlier). The present study was conducted in the Mekelle

Outlier, an almost circular area of Mesozoic sediments

extending for about 8,000 km2.

The geology of the Mekelle area consists mainly of the

Agula Shale Formation which unconformably overlies the

Antalo Formation. The Antalo Limestone Formation is a

dominantly calcareous and marl succession formed during a

transgression phase in the Jurassic period. Beyth (1971,

1972) records it thickens progressively towards the Red Sea.

The Agula Shale Formation consists mainly of a number of

cyclic facies with variable thicknesses, largely represented

by limestone, shale and marl deposited during the Jurassic. It

contains numerous dolerite sills and dykes. Mengesha et al.

(1996) suggested it was formed in a lagoon during a

regression phase of the Jurassic sea. Regionally, the Agula

Shale Formation is overlain either by the Amba Aradom

Sandstone Formation or by Tertiary volcanic rocks.

Levitte (1970) pointed out that faulting and tectonic

movements control the structure of the Mekelle area. The

basement structures have the same strike, trending N25�E,

with small local deviations. The sediments in the plateau

are planar and sub-horizontal with dips varying between

30� and 90� to the northeast. The inclination is due to the

slight tilting of blocks hinged on north-westerly trending

faults.

Three main fault systems exist in the area; two limited to

the Ethiopian plateau and one to the rift of the Danakil

depression (Fig. 2). The fault systems of the plateau are

normal to each other. One system is characterized by

vertical faults up to 40 km long that cut the entire sedi-

mentary section and trend N25�E. The second system of

faults generally strikes in a WNW–ESE direction. Beyth

(1971) also studied the structure and tectonics of the sed-

imentary rocks in the Mekelle Outlier and in the escarp-

ment. By giving less emphasis to the N25�E trending fault

system of Levitte (1970), Beyth identified only two main

fault trends: the WNW fault belts (Wukro, Mekelle,

Chelekot and Fuicea Mariam) and the Rift Valley Fault

forming the escarpment and the Danakil depression.

The longest of these faults in the area is the Mekelle

Fault which passes along the periphery of the proposed

Geba dam site, Ethiopia 341

123

reservoir northeast of the dam site and forms a 65 km long

escarpment. Faulting brings the lowermost part of the

Antalo Formation against its uppermost part and in places

against the overlying Agula Formation, near the city of

Mekelle, implying a throw of at least 400 m (Bosellini

et al. 1997; Levitte 1970).

Methodology

The fieldwork included engineering geological mapping,

discontinuity surveying, core drilling, Lugeon tests and the

acquisition of 38 soil and 10 rock samples from the site of

the proposed dam. The mapping was undertaken along

north–south lines at different intervals in order to intersect

the different geological units and was complemented by air

photograph interpretation. The discontinuity survey inclu-

ded not only slope face and scanline mapping but also

information from the rock cores. The orientation data were

analyzed using a computer program based on equal area

stereographic projection (Rockworks15, Rockware 2010),

in the form of contoured pole and rose diagrams. Quanti-

tative description of discontinuities including orientation,

spacing, persistence, roughness, aperture and filling were

determined in accordance with ISRM (1981).

After detailed geological and geomorphological field-

work, a systematic geotechnical drilling campaign was

planned and implemented with a wireline NQ type

(75.7 mm diameter) diamond bit, along the dam axis and

reservoir site. Twenty boreholes were drilled to depths of

between 30 and 120 m, with a total core length of 1,283 m.

Logging and rock quality designation (RQD) measurement

was based on ISRM (1981), care being taken to separate

the artificial fractures created during the drilling.

Seventy-seven water pressure or Lugeon tests were car-

ried out using expandable double packers. The test section

length was varied from 1.5 to 5 m (with an average of 4 m)

and the test pressure adjusted for each section to take account

of the depth and the nature/type of rock to avoid hydraulic

fracturing and jacking. The Lugeon values were calculated

and the type of flow and behaviour were determined for each

of the test sections. According to Houlsby (1976) and Camilo

Quinones-Rozo (2010), laminar flow occurs when the Lu-

geon value of a rock mass is independent of the test pressure

while turbulent flow is indicated by an inverse relationship

between water pressure and Lugeon value. In the case of

dilation, similar hydraulic conductivities are observed at low

and medium pressures and much greater values at the max-

imum pressure. An increase in Lugeon value regardless of

the changes in water pressure is an indication of washout,

while a continuous decrease in Lugeon value regardless of

the changes in water pressure suggests void filling. A total of

77 pressure tests (63 on the dam site and 14 on the site of the

reservoir) were performed in 18 boreholes (12 on the dam

site and 6 on the site of the reservoir).

Results and discussion

At the initial stage of the investigation the geology of the

site was described in the field by conventional field

techniques.

The site is characterized by Mesozoic sedimentary

sequences, typically limestone, limestone-shale intercalations,

Fig. 2 Fault map of Mekelle

Outlier (modified after Levitte

1970). Small rectangle shows

location of Geba dam site

342 G. Berhane, K. Walraevens

123

shale and travertine underlying unconsolidated colluvial and

alluvial deposits (Fig. 3). Thick fractured limestone covers

the left and right abutments of the dam site and extends

toward the reservoir on the upstream side, forming steep

cliffs. The uppermost part is dark grey to black in color,

finely crystalline, compacted, and fractured with parallel

horizontal bedding. Towards the foot of the cliff, thin beds of

fossiliferous limestone were observed in the massive, dark

brown rocks.

The limestone-shale intercalation unit is covered on both

abutment slopes by colluvial deposits and in the central

foundation by alluvial soils. This unit was exposed on the

southeastern part of the dam site (Fig. 3) and extended both

downstream and upstream. Its variable colour (light yel-

lowish to greenish grey) and its fissility are typical features

of this unit. The alternating beds form gentle slopes due to

their low resistance to weathering and erosion compared

with the thick limestones which form the cliffs.

The third rock unit is weak, yellowish, weathered, fissile

shale, which is only exposed on the northwest side of the

dam site. In places, thin beds of jointed limestone occur.

The travertine is whitish in color, generally porous,

highly weathered and with some traces of bedding and

lamination. In places, it shows some interconnected cavi-

ties which would lead to excessive leakage.

The central foundation (lower valley floor) is covered by

up to 20 m of thick alluvial deposits (gravel-sand mixtures

and clayey sandy soils). In places, the coarser river deposits

are cemented by calcite while the fine soils form a flat

topography, used as farm land. The distributions of these

deposits are good indicators of morphological changes

associated with past meanders of the Geba River.

Table 1 presents the results of unconfined compression

strength testing of rocks from the dam site. The liquid limit of

the alluvial deposit varied from 0 or non-plastic to 66 % with

an average value of 26 %; the plastic limit ranged from 0 or

non-plastic to 45 % with an average value of 13 %. The

dominant soil types are clayey/silty sand (SC, SM) and silty/

clayey gravel (GM, GC). Of the 38 samples tests, some 37 %

were found to be non-plastic. The permeability varied

between 10-5 and 10-3 cm/s, i.e. the material was of med-

ium permeability. As a consequence, it was proposed that the

alluvial deposit should be removed or cut-off before con-

struction of the dam, to control leakage, prevent uplift

pressures and instability of the dam and/or piping problems.

Discontinuity survey

Joints in outcrops of the abutment slope were studied based

on slope face and scanline mapping. Figure 4 shows a plot

of contoured pole concentrations and a rose diagram of 118

discontinuity measurements from both abutments. Mea-

surements were taken systematically with special emphasis

on joints with a favourable orientation for leakage with

respect to the alignment of the dam axis and reservoir

configuration. Discontinuity sets were identified visually in

Fig. 3 Simplified engineering geological map of Geba dam site. Boreholes and planned dam-axis are indicated

Geba dam site, Ethiopia 343

123

the field and from the rose diagram. Figure 4 shows three

dominant sets of discontinuities characterize both abut-

ments, NW–SE (J1), N–S (J2) and NE–SW (J3), which are

mainly vertical to sub-vertical (tectonic joints); in addition

bedding plane, J4, is horizontal. The discontinuities are

generally open, smoothly undulating and of low to high

persistent within the extent of the exposed surfaces, although

most discontinuities extend beyond the exposure limits

suggesting they are interconnected or intersect each other.

The average spacing of the discontinuity sets ranges from 0.5

to 3 m with an average spacing of about 1.5 m (Table 2).

Borehole logs

Careful visual observation and logging shows that the geo-

logical/geotechnical successions of the dam site and reser-

voir are very erratic, attributed to complex sedimentation

processes and subsequent tectonic and intrusive activities.

The core samples studied provided a picture of the cyclic

nature of the strata, the variation in fracture intensity, the

thickness of the layers and the overall engineering properties

of the material (e.g. strength and permeability).

The central river valley is covered by alluvial deposits

(gravel-sand mixtures and clayey sand, Figs. 3, 5a, 6) with

a thickness of up to 20 m, underlain by alternating frac-

tured and bedded limestone-shale intercalations (Figs. 5b,

6). These are extremely variable in thickness and character

and extend up to more than 120 m below ground level. The

geological sequences of the left and right abutments are

similar to those of the central foundation, except that there

is no alluvial overburden in the central area although in

places, thin layers of colluvial deposits and thick limestone

overlie the limestone-shale intercalation unit (Fig. 5).

Rock quality designation (RQD) values

The rock quality designation (RQD) value is defined as the

sum of the core sticks in excess of 10 cm long, expressed

Table 1 Unconfined

compressive strength of rocks

from Geba dam site

Location Depth interval

(m)

Rock type Unit weight

(kg/m3)

Unconfined

compressive

strength (MPa)

Left abutment (BH-03) 11.08–11.53 Limestone 2,730 22.9

109.79–109.95 Shale 2,576 12.04

Central foundation (BH-02) 25–25.16 Shale 2,422 12

70.83–71.16 Gypsum 2,759 25.4

89.15–89.35 Limestone 2,855 51

Right abutment (BH-01) 6.63–7.11 Limestone 2,854 26.5

71.4–71.61 Shale 2,841 13.3

Right abutment (BH-07) 64.83–65.11 Limestone 2,629 53.5

70.72–70.96 Shale 2,454 14.4

119–119.43 Gypsum 3,107 31.3

Fig. 4 Stereographic projection contoured pole and rose diagram of discontinuity strikes from both abutments

344 G. Berhane, K. Walraevens

123

as a percentage of the total length of core run (Deere 1964).

RQD has been used in many dam designs as a first rock

mass quality assessment parameter (Ez Eldin et al. 2007;

Ghazifard et al. 2006; Ozsan and Akin 2002). It is also

considered as an index of rock quality (Deere and Deere

1988) for preliminary assessment. Deere (1964) attempted

to find a relationship between the numerical intensity of

discontinuities to the rock mass quality and the significance

of this and its effect on the deformability of the rock mass.

He concluded that maintaining a consistent standard of

drilling, the percentage of solid core recovered depends on

the strength and number of discontinuities in the rock mass.

Uromeihy and Farrokhi (2011) pointed out that RQD has

some limitations for thinly layered rocks.

Plots of the RQD records and their mean at different parts

of the dam site are illustrated in Fig. 7. It is interesting that in

all cases the RQD value is extremely variable without any

clear trend with depth. In such circumstances it is very dif-

ficult and would generally be misleading to rely on RQD

value in deciding on the groutability of the material and to

select depth sections for permeability or Lugeon tests.

Table 2 Characteristics of major discontinuity sets

Discontinuity set Average spacing (m) Average aperture (cm) Persistence (m) Roughness Weathering degree

NW–SE (J1) 1.4 13.6 1.5 Smooth, planar Slightly weathered

N–S (J2) 3.0 13.8 3.1 Smooth, undulating Fresh to moderately

NE–SW (J3) 1.2 3.0 2.3 Smooth, undulating Slightly weathered

Horizontal (J4) 0.5 4 14 Smooth, planar Slightly weathered

Fig. 5 Selected core from Geba dam site. a Alluvial deposit with

gravels and boulders (BH-2, depth interval: 0.00–4.00 m), b lime-

stone-shale intercalation (BH-02, depth interval: 9.75–14.70 m)

Fig. 6 Simplified geological cross-section along Geba dam axis (facing upstream)

Geba dam site, Ethiopia 345

123

Fig. 7 Values of RQD at different parts of the dam site: a left abutment; b, c central river valley and d right abutment

346 G. Berhane, K. Walraevens

123

Commonly, if the value of RQD is low, it would be assumed

that the rock is highly fractured and with high fissure flows

such that grouting would be required. However, in thinly

bedded sedimentary rocks, such as shale, this is not true. The

low RQD value for shale may be attributable to its inherent

weakness and its tendency to lose strength when exposed to

air and moisture. In general, as seen in Fig 7a, b, the shale and

jointed limestone layers had lower RQD values. Basic sta-

tistical analysis of the samples in the dam site showed that

51 % of RQD values would be classified as very poor to poor

and 49 % fair to excellent (Table 3).

Lugeon/pressure test results

The hydraulic conductivity of the rock mass at the dam site

was evaluated by conducting a Lugeon test campaign using

expandable double packers. The results are shown in

Table 4 and Fig. 8 illustrates the results for 4 selected

boreholes. As seen in Table 4,[62 % of the Lugeon values

were above 3, indicating grouting treatment is required

(Houlsby 1976, 1990). The distribution of Lugeon values

below 3 (impervious) is lower for both abutments than for

the central foundation, probably due to the presence of

thick fractured limestone in the abutments.

Many researchers reported that due to the overburden

effect, hydraulic conductivity decreases with depth (Lee and

Farmer 1993; Nappi et al. 2005). Although a simple

correlation between Lugeon values and depth for the dam

site showed a general reduction in Lugeon value with

increasing depth, there were many exceptions (Fig. 8). A

study on fractured granite by Hamm et al. (2007) also showed

an inconsistent relationship between hydraulic conductivity

and depth. In addition, analysis of drilling data showed that

there was no direct relationship between RQD value and

Lugeon value (Figs. 7, 8). In such conditions, if the depths of

sections for Lugeon tests are based on RQD or drilling log,

they are unlikely to indicate the actual permeability of the

rock mass as the RQD does not record the number, aperture

or connectivity of discontinuities which significantly affect

permeability. The problem is illustrated by a comparison of

Figs. 7a and 8a. As a depth of 21.7–22.9 m, the limestone has

a higher Lugeon value (53) and relatively higher RQD (50)

than at 41.8–46.8 m where the alternating beds of limestone

and shale have zero Lugeon and RQD values. Gunay and

Milanovic (2005) reported hydraulic conductivities ranging

from 0.0 to 50 Lugeons for limestone in southwest Turkey

while Nonveiller (1989) reported a Lugeon value ranging

from 0.0 to 180 for a reservoir constructed on karst limestone

in Croatia.

Type of flow of water

The result of flow type analysis conducted for the different

parts of the dam site and reservoir area are plotted in Fig. 9.

Table 3 Statistical distribution of number of RQD values in the different rock quality classes

RQD (%) Rock quality

(Deere and Deere 1988)

Left abutment Central foundation Right abutment Percentage distribution

(%)BH-3, 10 & 21 BH-2, 4, 5, 6, 8, 9,

16, 18 & 19

BH-1, 7 & 11

0–25 Very poor 82 34 21 35.58

25–50 Poor 18 30 14 16.1

50–75 Fair 18 48 22 22.86

75–90 Good 7 42 15 16.62

90–100 Excellent 9 21 4 8.83

Table 4 Statistical distribution of number of Lugeon values in the different permeability classes for the pressure tests executed on the dam site

Lugeon value

range

Classification (Ghafoori

et al. 2011)

Left

abutment

Central foundation Right abutment Total number

of tests

Percentage

distribution

(%)BH-03, 10

& 21

BH-02, 06, 08, 09,

16 & 18

BH-01, 07, 11 & 19

0–3 Impervious 6 12 6 24 38

3–10 Low permeability 2 3 3 8 13

10–30 Medium permeability 10 1 5 16 25

30–60 High permeability 4 3 5 12 19

[60 Very high permeability 0 1 2 3 5

Total number of tests 22 20 21 63

Geba dam site, Ethiopia 347

123

Fig. 8 Plot of Lugeon values versus depth for selected boreholes: a left abutment; b central foundation; c, d right abutment

348 G. Berhane, K. Walraevens

123

Of the 77 pressure tests undertaken, some 35 % indicated

turbulent flow, 22 % laminar flow; 16 % dilation; 12 %

washout and 4 % void filling. Only 12 % of the tests

showed no flow record. Compared with the central foun-

dation, the distribution of turbulent flow is high in both

abutments, indicating fast flow in wide, open discontinu-

ities or voids. In view of the geology of the area, this would

suggest that mainly the limestone rock mass has open and

large discontinuities or interconnected dissolution cavities

not necessarily apparent at the surface.

Foundation treatment and groutability

Water leakage from a dam is always a problem, and par-

ticularly so where the preservation of water is essential in

semi-dry areas such as northern Ethiopia. The geological

investigation, including the discontinuity surveys, drilling

(RQDs, borehole logs, etc.) and Lugeon tests, has indicated

treatment of the foundations will be necessary. In the

central foundation, up to 20 m of alluvial deposits will

need to be removed and replaced with compacted imper-

vious clay material. Houlsby (1990, 1976) and Uromeihy

and Farrokhi (2011) suggested that when the Lugeon val-

ues are between 3 and 10, a single row of grouting holes is

required, while with values of over 10, a grout curtain

should include three rows of grouting holes. The majority

(62 %) of the Lugeon values in the dam site were found to

be higher than 3, and of these 79 % were [10. The vari-

ation of Lugeon values with depth for both abutments and

the central foundation is shown in Fig. 10. It can be seen

that in general the Lugeon values decrease to less than 3 at

a depth of about 100 m for the left abutment, 35 m for the

central foundation and 60 m for the right abutment. As a

consequence, a grout curtain including two to three rows of

grouting holes was recommended to a depth of 100 m for

the left abutment, 35 m for the central foundation and 60 m

for the right abutment. Turbulent flow (Fig. 9) indicated

big open discontinuities or dissolution cavities. This sug-

gested that a coarse grout is essential. In addition, as the

rock mass quality was generally poor, injection at holes

should be carried out in stages, using an up-down method

in 3–5 m sections. The analysis of the core logs showed a

cyclic limestone-shale sequence in all the boreholes, hence

the determination of the grouting sections must be based on

the geological units, bearing in mind that the shale units

will not take grout.

Recommendations and conclusions

This study assessed the engineering geological character-

istics of the proposed Geba dam site with particular

emphasis on the hydraulic conductivity and groutability of

the materials. The rock mass at the dam site was a sequence

of cyclic limestone-shale intercalations of variable thick-

nesses and degrees of fracturing and was characterized by

both bedding and tectonic discontinuities. The RQD and

Lugeon values did not show any clear relationship, but as

many of the rocks with low RQDs had low Lugeon values,

the use of RQD as a parameter for the selection of Lugeon

test sections, is not applicable.

Water flow during the Lugeon tests was found to be

dominantly of turbulent type suggesting interconnected and

open discontinuity conditions at the dam site. About 62 %

of Lugeon values were found to be greater than 3, and of

these, 79 % had values greater than 10, indicating exces-

sive leakage through the rock foundations should be

expected.

The results of the discontinuity surveys, Lugeon tests and

drilling showed that the dam site was complex and needs

Fig. 9 Dominant water flow

type during pressure tests

conducted at Geba dam site and

reservoir

Geba dam site, Ethiopia 349

123

close consideration throughout the detailed design and

construction phases. A grout curtain with two to three rows of

grouting holes was recommended to a depth of 100 m for the

left abutment, 35 m for the central foundation and 60 m for

the right abutment. In addition, coarse grout should be

injected in stages, using an up-down method at 3–5 m

Fig. 10 Variability of Lugeon values with depth: a left abutment, b central foundation and c right abutment

350 G. Berhane, K. Walraevens

123

sections. It was recommended that the design and layout of

the grouting should be reviewed as more information

becomes available during design and construction phases.

Acknowledgments The authors are very grateful to the Department

of Earth Science of Mekelle University for providing logistic support

to conduct the fieldwork. Thanks also to the Flemish Interuniversity

Council—University Cooperation for Development (VLIR-UOS) for

the short research stay grant at Ghent University, Belgium, which

allowed the first author to prepare this article. We appreciate the data

provided by the Tigray Water Resource, Mines and Energy Bureau

and Addis Geosystems PLC and the opportunity given to the first

author to work with them during the investigation program. Thanks

are also given to the staff at the Laboratory for Applied Geology and

Hydrogeology of Ghent University for their assistance during the

work.

References

Abdulkadir M (2009) Assessment of micro-dam irrigation projects

and runoff predictions for ungauged catchments in northern

Ethiopia. PhD dissertation, Muenster University, Germany

Angulo B, Morales T, Uriarte JA, Antiguedad A (2011) Hydraulic

conductivity characterization of a karst recharge area using water

injection tests and electrical resistivity logging. Eng Geol

117:90–96

Berhane G (2010a) Geological, geophysical and engineering geolog-

ical investigation of a leaky micro-dam in the northern Ethiopia.

Agric Eng Int CIGR J 12(1):31–46

Berhane G (2010b) Engineering geological soil and rock character-

ization in the Mekelle town, northern Ethiopia: implications to

engineering practice. Momona Ethiop J Sci 2(2):64–86

Berhane G, Ayenew T (2010) Soil and rock characterization in the

Mekele area, northern Ethiopia. Int J Earth Sci Eng

3(6):762–774

Beyth M (1971) The geology of central and western Tigray.

Unpublished report, Ethiopian Institute of Geological Survey

(EIGS), Addis Ababa

Beyth M (1972) Paleozoic sedimentary basin of Mekelle Outlier,

northern Ethiopia. Am Assoc Pet Geol Bull 56:2426–2439

Bonacci O, Roje-Bonacci T (2008) Water losses from the Ricice

reservoir built in the Dinaric karst. Eng Geol 99:121–127

Bosellini A, Russo A, Fantozzi PL, Assefa G, Tadesse S (1997) The

Mesozoic succession of the Mekelle Outlier (Tigrai Province,

Ethiopia). Mem Sci Geol 49:95–116

Camilo Quinones-Rozo PE (2010) Lugeon test interpretation, revis-

ited. In: Proceedings of the 30th annual USSD conference

Sacramento, California, 12–16 Apr 2010, pp 405–414

Deere DU (1964) Technical description of cores for engineering

purposes. Rock Mech Eng Geol 1:18–22

Deere DU, Deere DW (1988) The rock quality designation (RQD)

index in practice. In: Kirakaldie L (ed) Rock Classification

systems for engineering purposes. ASTM special publication

984. American Society for Testing Materials, Philadelphia,

pp 91–101

Desta LT (2005) Reservoir siltation in Ethiopia: causes, source areas

and management options. PhD dissertation, University of Bonn,

Germany

Ez Eldin MAM, Huiming T, Bahwi NH, Faraw AG (2007)

Geological, soil and rock mass evaluation for proposed hydro-

electric power plant at Sennar Dam, Sudan. J Appl Sci

7(22):3477–3484

Foyo A, Tomillo C, Maycotte JI, Willis P (1997) Geological features,

permeability and groutability characteristics of the Zimapan

Dam foundation, Hidalgo State, Mexico. Eng Geol

46(2):157–174

Foyo A, Sanchez MA, Tomillo C (2005) A proposal for a Secondary

Permeability Index obtained from water pressure tests in dam

foundations. Eng Geol 77:69–82

Ghafoori M, Lashkaripour GR, Tarigh Azali S (2011) Investigation of

the geological and geotechnical characteristics of Daroongar

dam, Northeast Iran. Geotech Geol Eng 29:961–975. doi:10.

1007/s10706-011-9429-6

Ghazifard A, Heidari E, Hashemi M, Hangara A (2006) Evaluation of

engineering geological characteristics for the Kuhrang III dam

site, Iran. In: 10th IAEG international congress, 6–10 Sept 2006,

Nottingham, United Kingdom

Gonzalez-Quijano M (2006) Groundwater modelling of the Tsinkanet

catchment: a MODFLOW approach to evaluate the impact of

small reservoirs on groundwater recharge. M.Sc. thesis, Univer-

sity of Ghent and University of Brussels, Belgium

Goodman R, Moye D, Shalkwyk A, Javandel I (1965) Groundwater

inflow during tunnel driving. Bull Assoc Eng Geol 2:39–56

Gunay G, Milanovic P (2005) Karst engineering studies at the

Akkopru reservoir area, SW of Turkey. In: Proceedings of the

international conference and field seminars ‘‘water resources and

environmental problems in Karst’’, Belgrade and Kotor, Serbia

and Montenegro, pp 651–658

Gurocak Z, Alemdag S (2011) Assessment of permeability and

injection depth at the Atasu dam site (Turkey) based on

experimental and numerical analyses. Bull Eng Geol Environ.

doi:10.1007/s10064-011-0400-9

Hamm SY, Kim M, Cheong JY, Kim JY, Son M, Kim TW (2007)

Relationship between hydraulic conductivity and fracture prop-

erties estimated from packer tests and borehole data in a

fractured granite. Eng Geol 92:73–87

Haregeweyn N, Poesen J, Nyssen J, Verstraeten G, de Vente J, Govers

G, Deckers S, Moeyersons J (2005) Specific sediment yield in

Tigray-northern Ethiopia: assessment and semi-quantitative

modelling. Geomorphology 69:315–331

Heuer R (1995) A quantitative, empirical and theoretical approach on

water flow into tunnels. In: Rapid excavation and tunneling

conference, San Francisco, CA, 18–21 June

Houlsby AC (1976) Routine interpretation of the Lugeon water test.

Q J Eng Geol 9:303–313

Houlsby AC (1990) Construction and design of cement grouting: a

guide to grouting in rock foundation. Wiley, New Jersey

ISRM (1981) Suggested methods for the quantitative description of

discontinuities in rock masses. In: Barton N (ed) Rock charac-

terization, testing and monitoring. Pergamon, Oxford

Izharul H, Hashmi FAS (1983) Permeability tests at the Simly Dam

Project. Bull Eng Geol Environ 26–27(1):433–438

Kiraly L (1969) Anisotropy and heterogeneity of permeability in

fractured limestones. Eclogae Geol Helv 62(2):613–619

Kiraly L (1978) Definition of the hydrogeological unit. Bull Cent

Hydrogeol 2:83–216

Koutsoyiannis D (2011) Scale of water resources development and

sustainability: small is beautiful, large is great. Hydrol Sci J

56(4):553–575

Lee CH, Farmer I (1993) Fluid flow in discontinuous rocks. Chapman

& Hall, New York

Lee EJ, Schwab KJ (2005) Drinking water distribution systems in

developing countries. J Water Health 3(2):109–127

Levitte D (1970) The geology of Mekele (report on the geology of the

central part of sheet ND 37–11). Geological Survey of Ethiopia,

Addis Ababa

Lugeon M (1933) Barrages et Geologie. Dunod, Paris

Mengesha T, Tadiwos C, Workineh H (1996) Explanation of the

geological map of Ethiopia, scale 1:2,000,000. Ethiopian Insti-

tute of Geological Surveys Bull. No. 3

Geba dam site, Ethiopia 351

123

Mollah MA, Sayed SAS (1995) Assessment of in situ permeability

with emphasis on packer testing in Kuwait. Eng Geol

39:217–231

Morrow CA (2000) Permeability of deep drillhole core samples. In:

Proceedings of the international workshop on the Nojima fault

core and Borehole analysis, USGS

Mozafari M, Raeisi E, Zare M (2011) Water leakage paths in the

Doosti Dam, Turkmenistan and Iran. Environ Earth Sci. doi:10.

1007/s12665-011-1069-x

Nappi M, Esposito L, Piscopo V, Rega G (2005) Hydraulic

characterization of some arenaceous rocks of Molise (Southern

Italy) through outcropping measurements and Lugeon tests. Eng

Geol 81:54–64

Nedaw D, Walraevens K (2009) The positive effect of micro-dams for

groundwater enhancement: a case study around Tsinkanet and

Rubafeleg area, Tigray, northern Ethiopia. Momona Ethiop J Sci

1(1):59–73

Nonveiller E (1989) Grouting theory and practice. Elsevier,

Amsterdam

Ozsan A, Akin M (2002) Engineering geological assessment of the

proposed Urus Dam, Turkey. Eng Geol 66:271–281

Ozsan A, Karpuz C (1996) Geotechnical rock-mass evaluation of the

Anamur dam site, Turkey. Eng Geol 42:65–70

Rockware (2010) Rockworks15 manual, 3rd edn. Rockware, Inc.,

USA

Sharghi Y, Siahkoohi H, Alinia F, Moarefvan P (2010) Estimation of

Lugeon number at the abutments of Bakhtyari dam site using

seismic tomography. Aust J Basic Appl Sci 4(2):274–285

Snow DT (1968) Rock fracture spacing, openings and porosities.

J Soil Mech Found Eng 94(1):73–92

United Nations Environment Programme (2002) Vital water graphics:

an overview of the state of the world’s fresh and marine waters.

United Nations, Nairobi

United Nations Population Division (2002) World urbanization

prospects: the 2001 revision. United Nations, New York

United Nations Population Division (2010) World urbanization

prospects: the 2009 revision. United Nations, New York

Uromeihy A, Farrokhi R (2011) Evaluating groutability at the Kamal-

Saleh Dam based on Lugeon tests. Bull Eng Geol Environ.

doi:10.1007/s10064-011-0382-7

Walraevens K, Vandecasteele I, Martens K, Nyssen J, Moeyersons J,

Gebreyohannes T, De Smedt F, Poesen J, Deckers J, Van Camp

M (2009) Groundwater recharge and flow in a small mountain

catchment in northern Ethiopia. Hydrol Sci J 54(4):739–753

Wolela A (2008) Sedimentation of the Triassic–Jurassic Adigrat

Sandstone Formation, Blue Nile (Abay) Basin, Ethiopia. J Afr

Earth Sci 52:30–42

WWDSE (2007) Evaluation of Aynalem wellfield and selection of

prospective well fields around Mekelle Town for water supply

source. Unpublished technical report, Addis Ababa

Yamaguchi Y, Shibuichi H, Matsumoto N (1997) Permeability

evaluation of jointed rock masses using high viscosity fluid tests.

Int J Rock Mech Min Sci 34:344.e1–344.e15

352 G. Berhane, K. Walraevens

123